Antibiotic use numeracy: Developing the infectious numeracy test (INT).

Background: Inadequate numeracy is a barrier to patients' understanding of their health. No measure of numeracy related to antibiotic use is available.

Objective: to develop and validate a measure of numeracy of antibiotic use, and to make this measure available in the Arabic language and context.

Methods: Best practices for developing, translating and validating scales were consulted to develop an anonymous survey administered on SurveyMonkey. Ten open-ended questions for which participants provided an answer assessed basic numeracy covering fractions, percentages, proportionality, measurement, and estimation, contextualized to real life situations. A panel of five experts, and 10 cognitive debriefings provided face and content validity for the tool. Individuals ≥18 years, who had mastery of Arabic or English were included. Exploratory factor analysis was performed to characterize the psychometric properties of the items, using principal-component analysis. Convergent validity was established by comparing numeracy scores with the antibiotic knowledge scale scores. Differences in numeracy scores according to sociodemographic variables were tested using multivariate analyses of variance.

Results: Two hundred fifty-four responses were obtained, giving a response rate of 63.5%. A two-factor structure which explained 44.2% of the variance evolved: "Mathematical knowledge and problem-solving skills" and "Numeracy-related practices and experience". An acceptable reliability of Cronbach's alpha coefficient = 0.713. Higher numeracy scores were significantly correlated with the scores of the antibiotic knowledge (r = 0.205, p < 0.01) and factor one within the tool: "side effects and resistance" (r = 0.162, p = 0.05). Participants with University/College education presented higher levels of numeracy-related practices and experience than those with high-school education. Likewise, participants with medical insurance presented with higher numeracy scores than those without.

Conclusions: A valid and reliable measure of antibiotic numeracy in English and Arabic was created. The tool could identify patients with low numeracy who are candidates for targeted interventions and intensive education on appropriate antibiotic use.